The three components of this Program Project will use novel approaches to study the role of Pten as a tumor suppressor in melanoma development and progression. While about 30% of melanoma harbors inactive form of Pten, over 50% have been reported to carry a constitutively active form of PI3K/Akt, pointing to the importance of this pathway in melanomas. It is anticipated that the studies proposed in this application will provide important new mechanistic insight into the role of Pten/Akt signaling cascade in melanoma development and progression while identifying new pharmacological targets and the developing drug candidates with potent anti-tumor activity. The use of both cell culture and animal models will be essential to the success of these programmatic plans. The primary goals of Core C are to support all Projects as follows: 1) To provide proven tools and expertise in assessment of expression signatures using a series of antibodies against proteins involved in the Pten signaling pathway;2) to use monospecific commercial or "homemade" antibodies to perform an assessment of genes / proteins discovered and characterized in the course of this PPG. These genes and proteins will serve as potential markers for diagnosis and monitoring of Pten-inactive tumors. This analysis will be carried out on >30 melanoma tumor derived cell lines in which the status of Pten is known, and on TMAs of melanoma patients that are available to us as part of collaboration with David Rimm's group at Yale university;3) to provide expert pathological support for studies using experimental in vivo and in vitro models;4) to ensure that optimal biostatistical analysis is used to validate the diagnostic data that are obtained;5) To provide in vitro tumor cell models to serve the needs of all program components. These models include both 2-dimensional melanoma tumor cell monolayers, as well as 3-dimensional cultures of melanoma cell spheroids which in some cases may more faithfully mimic the in v/Vo tumor cell environment and its response to drug treatment;6) to provide in vivo tumor models utilizing subcutaneous human xenografts in nude mice (Pten-negative and reconstituted human melanoma cells of the UACC903 series, as well as WM35 and WM35/Akt over- expressing cells) that allow quantitative analysis of tumor growth and metastasis. These standardized models will allow reliable comparisons of drug testing and other types of data among programmatic components.